Boundary-guided cell alignment drives mouse epiblast maturation

MainDuring embryonic development and homeostasis, cells and tissues repeatedly break symmetry and form patterns. At the tissue scale, patterns can emerge through biochemical and mechanical interactions between cells or at the supracellular level. Morphogen signalling, for example, instructs the formation of tissue patterns with distinct material properties1,2,3,4. Although these mechanisms have been studied, the effect of tissue boundaries on tissue patterning remains less understood5,6,7,8, particularly when properties of these boundaries are heterogeneous. Recent studies have begun to investigate how interactions between cells and the extracellular matrix (ECM) deposited at tissue boundaries contribute to tissue formation9,10. However, specific mechanisms by which cell–ECM interactions influence cell arrangement, tissue patterning and their subsequent functional consequences remain elusive.Orientational order—a spatial pattern in living systems—emerges at various scales, from subcellular structures to whole organisms11,12,13. Directional cellular organization, analogous to nematic ordering in liquid crystals14, has been observed in two-dimensional (2D) cell cultures, such as neural progenitors15 and epithelial cells16. However, our understanding of three-dimensional (3D) cell arrangement in tissues remains limited, primarily due to the technical challenges associated with monitoring cellular dynamics within a 3D tissue undergoing pattern formation. Moreover, investigating the potential impact of cell–ECM interactions at the tissue boundary requires approaches to manipulate ECM deposition without disrupting overall tissue architecture.The early mouse embryo provides a model system to investigate the mechanisms of tissue pattern formation17,18,19. Mammalian embryos are derived from the epiblast (EPI) that forms in the blastocyst as an aggregate of non-polarized cells18,20,21,22,23. Upon implantation and by embryonic day 5.5 (E5.5), the EPI undergoes maturation, a developmental transition during which EPI cells elongate, acquire apico-basal polarity and arrange in a radial manner, while the EPI tissue transforms into the cup-shaped egg-cylinder structure17,24,25,26 (Fig. 1a). During this morphogenesis, the EPI is enveloped by two extraembryonic tissues: the primitive endoderm (PrE) (which differentiates into the visceral endoderm (VE)) and the polar trophectoderm (pTE) (which gives rise to the extraembryonic ectoderm (ExE)). These interactions establish two distinct tissue boundaries with potentially different properties. Indeed, the precise cellular and molecular mechanisms underlying EPI patterning remain elusive, largely due to limited access to the dynamic cellular processes occurring within the uterine tissue.Fig. 1: Progressive cell alignment and orientation to the boundary underlie EPI patterning.Full size imagea, Schematic (top) illustrating morphological changes at the cellular and tissue levels during the E4.5–E5.25 window, from blastocyst to cup-shaped egg-cylinder structure with elongated, radially oriented EPI cells (cyan). The EPI is surrounded by pTE/ExE (grey) and PrE/VE (red); other parts of the embryo are shown in light grey. Immunofluorescence images (first row) and the corresponding membrane segmentation (second row) of representative embryos from E4.5 to E5.25, stained for Oct3/4 (EPI; cyan) and cell membrane (orange), generated by combining E-cadherin and phalloidin signals. The third row shows the 3D visualization of segmented cell volumes viewed from a specific angle, with coordinate axes (arrows) indicating the x, y and z dimensions. The bottom row shows the cell long axes extracted by computing principal inertia vectors in three dimensions from the segmented volumes. n = 37 (E4.5), 26 (E4.75), 22 (E5.0) and 12 (E5.25) embryos segmented and analysed from at least three independent embryo recovery experiments. b, Heat map showing the normalized distribution of angles between the long axes of neighbouring cells, binned in 10° intervals (0°–90°). Data from E4.5 to E5.25 embryos are grouped by the EPI cell number in intervals of 15. Colour intensity represents normalized frequency within each group. Sample sizes by the EPI cell number: [15–29], n = 732 cells from 36 embryos; [30–44], 1,226 from 36; [45–59], 890 from 18; [60–74], 1,077 from 17; [75–89], 626 from 8; [90–104], 373 from 4. c,d, Angle measurement between the cell long axis and the normal vector to the tissue boundary, represented as violin plots with individual data points overlaid. Each plot shows the distribution of angles for EPI cells in contact with the ExE-boundary (c) and VE-boundary (d). Dot colours indicate the embryo stage. Data are grouped by the EPI cell number, with median values shown by red bars. Sample sizes for c: [15–29], n = 593 cells from 36 embryos; [30–44], 750 from 35; [45–59], 220 from 15; [60–74], 196 from 12; [75–89], 104 from 8; [90–104], 61 from 5. For d: [15–29], n = 618 cells from 36 embryos; [30–44], 951 from 35; [45–59], 682 from 15; [60–74], 649 from 12; [75–89], 540 from 8; [90–104], 446 from 5. Mann–Whitney U-test (two sided) without correction for multiple comparisons; each group compared with the reference group [15–29 cells]. ***P 0.05; *P < 0.05; **P < 0.01; ***P < 0.001.Materials availabilityAll unique/stable reagents generated in this study are available from the corresponding authors with a completed Materials Transfer Agreement.Reporting summaryFurther information on research design is available in the Nature Portfolio Reporting Summary linked to this article.